Bootstrapping social emotion classification with semantically rich hybrid neural networks

Xiangsheng LI, Yanghui RAO*, Haoran XIE, Raymond Yiu Keung LAU, Jian YIN, Fu Lee WANG

*Corresponding author for this work

Research output: Journal PublicationsJournal Article (refereed)

16 Citations (Scopus)

Abstract

Social emotion classification aims to predict the aggregation of emotional responses embedded in online comments contributed by various users. Such a task is inherently challenging because extracting relevant semantics from free texts is a classical research problem. Moreover, online comments are typically characterized by a sparse feature space, which makes the corresponding emotion classification task very difficult. On the other hand, though deep neural networks have been shown to be effective for speech recognition and image analysis tasks because of their capabilities of transforming sparse low-level features to dense high-level features, their effectiveness on emotion classification requires further investigation. The main contribution of our work reported in this paper is the development of a novel model of semantically rich hybrid neural network (HNN) which leverages unsupervised teaching models to incorporate semantic domain knowledge into the neural network to bootstrap its inference power and interpretability. To our best knowledge, this is the first successful work of incorporating semantics into neural networks to enhance social emotion classification and network interpretability. Through empirical studies based on three real-world social media datasets, our experimental results confirm that the proposed hybrid neural networks outperform other state-of-the-art emotion classification methods.

Original languageEnglish
Article number7953530
Pages (from-to)428-442
Number of pages15
JournalIEEE Transactions on Affective Computing
Volume8
Issue number4
Early online date19 Jun 2017
DOIs
Publication statusPublished - Oct 2017
Externally publishedYes

Fingerprint

Neural networks
Semantics
Speech recognition
Image analysis
Teaching
Agglomeration

Keywords

  • hybrid neural network
  • Social emotion classification
  • sparse encoding
  • transfer learning

Cite this

LI, Xiangsheng ; RAO, Yanghui ; XIE, Haoran ; LAU, Raymond Yiu Keung ; YIN, Jian ; WANG, Fu Lee. / Bootstrapping social emotion classification with semantically rich hybrid neural networks. In: IEEE Transactions on Affective Computing. 2017 ; Vol. 8, No. 4. pp. 428-442.
@article{45a11e227f4e44c78fc344c55ce87fc4,
title = "Bootstrapping social emotion classification with semantically rich hybrid neural networks",
abstract = "Social emotion classification aims to predict the aggregation of emotional responses embedded in online comments contributed by various users. Such a task is inherently challenging because extracting relevant semantics from free texts is a classical research problem. Moreover, online comments are typically characterized by a sparse feature space, which makes the corresponding emotion classification task very difficult. On the other hand, though deep neural networks have been shown to be effective for speech recognition and image analysis tasks because of their capabilities of transforming sparse low-level features to dense high-level features, their effectiveness on emotion classification requires further investigation. The main contribution of our work reported in this paper is the development of a novel model of semantically rich hybrid neural network (HNN) which leverages unsupervised teaching models to incorporate semantic domain knowledge into the neural network to bootstrap its inference power and interpretability. To our best knowledge, this is the first successful work of incorporating semantics into neural networks to enhance social emotion classification and network interpretability. Through empirical studies based on three real-world social media datasets, our experimental results confirm that the proposed hybrid neural networks outperform other state-of-the-art emotion classification methods.",
keywords = "hybrid neural network, Social emotion classification, sparse encoding, transfer learning",
author = "Xiangsheng LI and Yanghui RAO and Haoran XIE and LAU, {Raymond Yiu Keung} and Jian YIN and WANG, {Fu Lee}",
year = "2017",
month = "10",
doi = "10.1109/TAFFC.2017.2716930",
language = "English",
volume = "8",
pages = "428--442",
journal = "IEEE Transactions on Affective Computing",
issn = "1949-3045",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "4",

}

Bootstrapping social emotion classification with semantically rich hybrid neural networks. / LI, Xiangsheng; RAO, Yanghui; XIE, Haoran; LAU, Raymond Yiu Keung; YIN, Jian; WANG, Fu Lee.

In: IEEE Transactions on Affective Computing, Vol. 8, No. 4, 7953530, 10.2017, p. 428-442.

Research output: Journal PublicationsJournal Article (refereed)

TY - JOUR

T1 - Bootstrapping social emotion classification with semantically rich hybrid neural networks

AU - LI, Xiangsheng

AU - RAO, Yanghui

AU - XIE, Haoran

AU - LAU, Raymond Yiu Keung

AU - YIN, Jian

AU - WANG, Fu Lee

PY - 2017/10

Y1 - 2017/10

N2 - Social emotion classification aims to predict the aggregation of emotional responses embedded in online comments contributed by various users. Such a task is inherently challenging because extracting relevant semantics from free texts is a classical research problem. Moreover, online comments are typically characterized by a sparse feature space, which makes the corresponding emotion classification task very difficult. On the other hand, though deep neural networks have been shown to be effective for speech recognition and image analysis tasks because of their capabilities of transforming sparse low-level features to dense high-level features, their effectiveness on emotion classification requires further investigation. The main contribution of our work reported in this paper is the development of a novel model of semantically rich hybrid neural network (HNN) which leverages unsupervised teaching models to incorporate semantic domain knowledge into the neural network to bootstrap its inference power and interpretability. To our best knowledge, this is the first successful work of incorporating semantics into neural networks to enhance social emotion classification and network interpretability. Through empirical studies based on three real-world social media datasets, our experimental results confirm that the proposed hybrid neural networks outperform other state-of-the-art emotion classification methods.

AB - Social emotion classification aims to predict the aggregation of emotional responses embedded in online comments contributed by various users. Such a task is inherently challenging because extracting relevant semantics from free texts is a classical research problem. Moreover, online comments are typically characterized by a sparse feature space, which makes the corresponding emotion classification task very difficult. On the other hand, though deep neural networks have been shown to be effective for speech recognition and image analysis tasks because of their capabilities of transforming sparse low-level features to dense high-level features, their effectiveness on emotion classification requires further investigation. The main contribution of our work reported in this paper is the development of a novel model of semantically rich hybrid neural network (HNN) which leverages unsupervised teaching models to incorporate semantic domain knowledge into the neural network to bootstrap its inference power and interpretability. To our best knowledge, this is the first successful work of incorporating semantics into neural networks to enhance social emotion classification and network interpretability. Through empirical studies based on three real-world social media datasets, our experimental results confirm that the proposed hybrid neural networks outperform other state-of-the-art emotion classification methods.

KW - hybrid neural network

KW - Social emotion classification

KW - sparse encoding

KW - transfer learning

UR - http://www.scopus.com/inward/record.url?scp=85021784569&partnerID=8YFLogxK

U2 - 10.1109/TAFFC.2017.2716930

DO - 10.1109/TAFFC.2017.2716930

M3 - Journal Article (refereed)

AN - SCOPUS:85021784569

VL - 8

SP - 428

EP - 442

JO - IEEE Transactions on Affective Computing

JF - IEEE Transactions on Affective Computing

SN - 1949-3045

IS - 4

M1 - 7953530

ER -